Understanding the role of psychological factors in long COVID: a network analysis approach

Abstract

Long COVID (LC) is a heterogeneous, multisystem condition that persists beyond the acute phase of SARS-CoV-2 infection. Psychological symptoms are highly prevalent and may influence the course and severity of LC. However, their specific role within the broader symptom structure remains insufficiently understood. This study applied a psychological network approach to examine how psychological factors contribute to the overall symptom structure of LC and to identify central and bridging variables that may serve as promising targets for intervention. A sample of 283 individuals with LC (nfemale = 235, nmale = 47, ndiverse = 1; age: M = 39.48, SD = 13.29) completed an online survey assessing post-viral physical symptoms and psychological factors, including depression, anxiety, COVID-19-related traumatic stress, and lack of self-efficacy. A regularized partial correlation network was estimated based on ten variables. The network revealed a dense degree of connectivity, with psychological factors integrated into the broader symptom structure. Depression emerged as the most central variable. Cardiovascular and respiratory symptoms, neurological symptoms, and depression served as key bridge variables. Lack of self-efficacy showed moderate associations with COVID-19-related traumatic stress and anxiety. Female gender was linked to greater gastrointestinal symptom burden, while older age was associated with more pronounced cardiovascular and respiratory symptoms. This study underscores the central role of psychological factors-particularly depression-as key targets for intervention in LC. By advancing the understanding of factors shaping health outcomes in LC, our findings support the integration of psychological approaches into the clinical management of affected individuals.